KR20210106903A - Position measurement apparatus, overlay inspection apparatus, position measurement method, imprint apparatus, and article manufacturing method - Google Patents

Abstract

Provided is a position measuring device for reducing a measurement error of an object. The position measuring device for measuring the position of an object includes: an illumination unit configured to illuminate the object with illumination light including light of a first wavelength and light of a second wavelength different from the first wavelength; a measurement unit configured to measure the position of the object by detecting light from the object illuminated with the illumination light; and a control unit configured to adjust a ratio of the light intensity of the second wavelength to the light intensity of the first wavelength so that a measurement error that varies depending on the position of the object in the measurement unit can be reduced.

Description

POSITION MEASUREMENT APPARATUS, OVERLAY INSPECTION APPARATUS, POSITION MEASUREMENT METHOD, IMPRINT APPARATUS, AND ARTICLE MANUFACTURING METHOD}

The present invention relates to a position measuring apparatus, an overlap inspection apparatus, a position measuring method, an imprint apparatus, and an article manufacturing method.

In recent years, a microfabrication technique for transferring a microstructure on a mold to a workpiece such as a semiconductor, glass, resin, or metal has been developed and attracts attention. Because these techniques are implemented with a resolution of the order of a few nanometers, they are referred to as nano-imprinting, nano-embossing, and the like, and they can collectively process three-dimensional structures at the wafer level in addition to the fabrication of semiconductors. They are expected to be applied in a wide range of fields as manufacturing technologies for optical devices such as photonic crystals, micro total analysis systems (μ-TAS), and biochips. In relation to such nano-imprinting, for example, a case in which an optical imprinting method is used in a semiconductor manufacturing technology will be described below.

First, a resin layer made of a photocurable resin is formed on a substrate (wafer) that is an object to be measured. Then, the mold in which the desired concave-convex structure is formed is brought into contact with the resin layer and pressed. Thereafter, the photocurable resin is cured by irradiating (illuminating) ultraviolet rays, and the concave-convex structure is transferred to the resin layer. Further, a desired structure is formed on the substrate by etching or the like using this resin layer as a mask. However, in manufacturing such a semiconductor, the mold and the substrate need to be aligned. For example, in a recent environment in which process rules for semiconductors require 100 nm or less, the allowable range of alignment errors due to devices is becoming strict enough to be several nm to several tens of nm.

In relation to such an alignment method, for example, in the specification of US Patent No. 6696220, a method of performing alignment in which a mold and a substrate are brought into contact with each other with a resin interposed therebetween is proposed. In the method employed by this, first, a photocurable resin is selectively applied to portions other than the alignment marks provided on the substrate. Then, the substrate is moved to a position facing the mold. In this state, the distance between the mold and the workpiece is shortened so that the mold is brought close to the substrate to a height such that the alignment mark is not buried in the resin. In this state, alignment is performed, and then final pressing is performed.

In the relative alignment, the measurement is performed after the mold and the substrate are brought into contact with each other by the resin therebetween. For this reason, at the time of measurement, resin is filled in the mark of a mold. Therefore, when the material of the mold and the resin have similar physical property values, there is a problem that the mold mark may not be seen.

When detecting the alignment mark, if there is an asymmetric processing error during the fabrication of the wafer, an error occurs in the measurement value. In order to reduce the error, for example, in Japanese Patent Laid-Open No. 2004-117030, the same mark is measured under different conditions (wavelength, polarization, etc.), and the amount of position shift is calculated using the wavelength having the highest contrast.

Observe the alignment marks patterned on the wafer to measure the wafer position. When an asymmetric processing error (shift due to the wafer) occurs in the wafer surface during the fabrication of the wafer, the measurement value is deceived upon observation of the alignment mark, resulting in measurement error. When a large measurement error occurs, the error causes a defect in the overlap exposure. Therefore, it is necessary to reduce the measurement error.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a position measuring device in which, for example, a measurement error of an object is reduced.

According to one aspect of the present invention, there is provided a position measuring device for measuring the position of an object. The position measuring device includes an illumination unit configured to illuminate an object with an illumination light including light of a first wavelength and light of a second wavelength different from the first wavelength, by detecting light from the object illuminated with the illumination light. a measurement unit configured to measure the position of the object; and a control unit configured to adjust a ratio of the light intensity of the first wavelength to the light intensity of the second wavelength such that a measurement error that varies depending on the position of the object in the measurement unit is reduced.

Additional features of the present invention will become apparent from the following description of exemplary embodiments with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows an example of the structure of the alignment light source which concerns on Example 1. FIG.
Fig. 2 is a diagram showing an example of the device configuration of the imprint apparatus according to the first embodiment.
3 is a diagram showing an example of the configuration of the measurement optical system according to the first embodiment.
4 is a diagram showing an example of the configuration of the measurement optical system according to the first embodiment.
It is a figure which shows an example of a structure of an alignment light source.
Fig. 6 is a diagram showing the pupil distribution of the measurement optical system according to the first embodiment.
7A to 7D are diagrams showing alignment marks for generating moire fringes.
8A to 8D are views showing alignment marks according to the first embodiment.
9A and 9B show the shape of a cross section of a pattern according to Example 1. FIG.
10A and 10B show a simulation model according to Example 1 and the results thereof.
11 is an operation sequence diagram of the imprint apparatus according to the first embodiment.
12 is a diagram showing a shift of a mark position according to the first embodiment.
13A and 13B are diagrams showing the synthesis of a lamp light source according to the first embodiment.
Fig. 14 is an example of arrangement of marks in a field of view according to the first embodiment.
15A and 15B are diagrams showing measurement errors occurring in the wafer surface according to the first embodiment.
16A and 16B are diagrams showing other measurement errors occurring in the wafer surface according to the first embodiment.
17A and 17B are diagrams showing other measurement errors occurring in the wafer surface according to the first embodiment.
18 is an example of a layout within a wafer surface for measuring a measurement error in a radial direction of a wafer with respect to a plurality of wavelengths according to Example 1. FIG.
19 is an example of a layout within a wafer surface for measuring measurement errors with respect to a plurality of wavelengths according to Example 1. FIG.
20 is an example of a layout within a wafer surface for measuring metrology errors with respect to a plurality of wavelengths according to Example 1. FIG.
21A and 21B are sequence diagrams for measuring a measurement error within a wafer surface with respect to a plurality of wavelengths according to Example 1. FIG.

EMBODIMENT OF THE INVENTION Hereinafter, with reference to an accompanying drawing, advantageous embodiment of this invention is demonstrated using an Example. In each drawing, the same reference numerals are assigned to the same members or elements, and overlapping descriptions are omitted or simplified.

(First embodiment)

FIG. 2 is a diagram showing the configuration of the imprint apparatus 1 of the first embodiment. This imprint apparatus 1 is used for the manufacture of devices such as semiconductor devices, and uses a mold (mold or mask) 7 to mold an uncured resin (resist) 9 on a wafer (substrate) as a processing target. , a device for forming (transferring) a pattern of resin 9 on a substrate. Here, the resin 9 is a resin which is hardened|cured by an ultraviolet-ray etc., for example. The imprint apparatus 1 of Example 1 employs a photocuring method. In addition, in the following drawings, X and Y axes orthogonal to each other in a plane parallel to the mold 7 and the wafer 8 are set, and the Z axis is set in a direction perpendicular to the X and Y axes. The imprint apparatus 1 includes a UV irradiation unit 2 , a measurement optical system 3 , a mold holding unit 4 , a wafer stage 5 , an application unit 6 , an acquisition unit (not shown), and A control unit (12) is provided.

The UV irradiation unit 2 is configured to irradiate the mold 7 with ultraviolet rays in order to harden the resin 9 after the mold pressurization process for bringing the mold 7 and the resin 9 on the wafer 8 into contact with each other. it is a device This UV irradiation unit 2 is composed of a light source (not shown), and a plurality of optical elements for uniformly irradiating ultraviolet rays emitted from the light source in a predetermined shape to an uneven pattern 7a, which will be described later, which is a surface to be irradiated. do. In particular, it is preferable that the irradiation area (irradiation range) of light by the UV irradiation unit 2 is approximately equal to the surface area of the concave-convex pattern 7a or slightly larger than the surface area of the concave-convex pattern 7a. This is to suppress the expansion of the mold 7 or wafer 8 due to the heat accompanying the irradiation and the occurrence of misalignment or distortion in the pattern transferred to the resin 9 by minimizing the irradiation area. . In addition, ultraviolet rays reflected by the wafer 8 or the like reach the application unit 6 (to be described later) and cure the resin 9 remaining in the discharge portion of the application unit 6, whereby the subsequent application unit ( It is also to prevent an abnormality from occurring in the operation of 6). Here, in relation to the light source, for example, a high-pressure mercury lamp, various excimer lamps, excimer lasers, or light emitting diodes may be employed. This light source is appropriately selected according to the characteristics of the resin 9 (light-receiving body). However, Example 1 is not limited to the type, number, wavelength, and the like of the light sources.

The measurement optical system (illumination unit and measurement unit) 3 optically detects the mold mark 10 disposed on the mold 7 and the wafer mark 11 disposed on the wafer 8 to determine the relative positions of both marks. It is an optical system for measuring. Further, the measurement optical system 3 functions as a part of an illumination unit configured to illuminate an object such as the wafer 8 by the illumination light from the alignment light source 23 . Further, the measurement optical system 3 also functions as a part of a measurement unit configured to measure the relative position of the object by detecting light from the object illuminated with the illumination light. Further, the measurement optical system 3 is arranged so that its optical axis is perpendicular to the mold 7 or the wafer 8 . Further, the measurement optical system 3 is configured to be able to be driven in the X-axis direction and the Y-axis direction depending on the position of the mold mark 10 or the wafer mark 11 . Further, the measurement optical system 3 is configured to be able to be driven also in the Z-axis direction in order to align the focus of the optical system to the position of the mold mark 10 or the wafer mark 11 . The driving of the wafer stage 5 or the magnification correction mechanism is controlled based on the information on the relative positions of the mold 7 and the wafer 8 measured by the measurement optical system 3 . The measurement optical system 3, the mold mark 10, and the wafer mark 11 (alignment mark) will be described in detail below.

The mold holding unit 4 is a mold holding unit that pulls and holds the mold 7 by vacuum suction force or electrostatic force. The mold holding unit 4 includes a mold chuck (not shown) and a mold drive that drives the mold chuck in the Z-axis direction in order to press the mold 7 against the resin 9 applied on the wafer 8 . includes instruments. Further, the mold holding unit 4 also includes a mold magnification correction mechanism for correcting the distortion of the pattern transferred to the resin 9 by deforming the mold 7 in the X-axis direction and the Y-axis direction. Each operation of mold pressing and mold separation in the imprint apparatus 1 can be realized by moving the mold 7 in the Z direction in this way. However, for example, the above operation can be realized by moving the wafer stage 5 (wafer 8) in the Z direction, or it can move both of them.

The wafer stage 5 is a wafer holder (substrate holder) that holds the wafer 8 using vacuum suction in a manner capable of moving, for example, in the XY plane.

The application unit (dispenser) 6 is an application unit configured to apply a part on the wafer 8 with a resin (uncured resin) 9 . Here, for example, the resin 9 is a photocurable resin having a property of being cured when receiving ultraviolet light, and is appropriately selected according to the type of the semiconductor device. The application unit 6 is not installed inside the imprint apparatus 1 as shown in FIG. 2, but an external application apparatus is separately prepared and the resin 9 is applied to the wafer 8 by this application apparatus in advance. can have a configuration realized by introducing into the imprint apparatus 1 . According to this configuration, since the application step inside the imprint apparatus 1 is eliminated, the processing in the imprint apparatus 1 can be accelerated. Moreover, since the application unit 6 is no longer required, the manufacturing cost in the imprint apparatus 1 as a whole can be suppressed.

Further, the mold 7 is a mold in which a predetermined pattern (for example, an uneven pattern 7a such as a circuit pattern) is formed in a three-dimensional shape on a surface facing the wafer 8 . The material of the mold 7 is quartz or the like capable of transmitting ultraviolet rays. In addition, for example, the wafer 8 is a target body made of single crystal silicon, and the target surface is coated with a resin 9 molded by a mold 7 . An acquisition unit (not shown) acquires information of a ratio (intensity ratio) of light intensities of at least two or more types of adjusted wavelengths, which will be described later.

The control unit 12 controls the UV irradiation unit 2 , the measurement optical system 3 , the mold holding unit 4 , the wafer stage 5 , and the application unit 6 . The control unit 12 may be composed of, for example, a field programmable gate array (FPGA), a computer having an embedded program, or a combination of all or part thereof. FPGAs may include programmable logic devices (PLDs) or application specific integrated circuits (ASICs). The control unit 12 has a built-in memory, a built-in CPU as a computer, and a control configured to execute various operations of the entire apparatus based on, for example, relational expressions, parameters, or computer programs stored (saved) in the memory. function as a unit. Further, the position measurement device is constituted by the control unit 12 , the alignment light source 23 , the measurement optical system 3 , and the like.

Next, the imprint process performed by the imprint apparatus 1 will be described. First, the wafer 8 is transferred to the wafer stage 5 by a substrate transfer unit (not shown), and the wafer 8 is loaded and fixed. Then, the wafer stage 5 is moved to the application position of the application unit 6 . Then, as an application step, the application unit 6 applies the resin 9 to a predetermined pattern formation area (shot area) of the wafer 8 (application step). Then, the wafer stage 5 is moved so that the coated surface on the wafer 8 is located directly under the mold 7 . Then, the mold driving mechanism is driven to perform mold pressurization to press the mold 7 against the resin 9 on the wafer 8 (mold pressurization). At this time, the resin 9 flows along the concave-convex pattern 7a formed in the mold 7 by the mold pressurization of the mold 7 . Further, in this state, the relative positions of the mold mark 10 and the wafer mark 11 arranged on the wafer 8 and the mold 7 are detected (measured) by the measurement optical system 3, and the wafer stage 5 ) to align the mold pressing surface of the mold 7 and the coated surface of the wafer 8 . Further, a magnification correction or the like is performed on the mold 7 by a magnification correction mechanism (not shown). Here, the wafer stage (alignment unit) 5 functions as an alignment unit configured to align the mold 7 and the wafer 8 . At a stage in which the flow of the resin 9 to the concave-convex pattern 7a, the alignment between the mold 7 and the wafer 8, correction of the magnification of the mold, etc. have been sufficiently performed, the UV irradiation unit 2 Ultraviolet rays are irradiated to the mold 7 from the rear surface (upper surface), and the resin 9 is cured using the ultraviolet rays transmitted through the mold 7 (curing step). At this time, the measurement optical system 3 is configured to be retracted so as not to block the optical path of the ultraviolet rays. Then, the mold driving mechanism is driven again to perform mold separation for separating the mold 7 from the wafer 8 (mold separation step). Thereby, the uneven pattern 7a of the mold 7 is transferred (formed) onto the wafer 8 (pattern forming step).

Subsequently, the details of the mold mark 10 and the wafer mark 11 for alignment arranged on the measurement optical system 3 and the mold 7 and the wafer 8, respectively, will be described. 3 is a diagram showing an example of the configuration of the measurement optical system 3 of the first embodiment.

The measurement optical system 3 is composed of a detection optical system 21 and an illumination optical system 22 . The detection optical system 21 captures an image of an interference fringe (moire fringe) generated by interference between the rays of the diffracted light from the wafer mark 11 and the mold mark 10 illuminated by the illumination optical system 22 to the imaging device ( 25) image above. The detection optical system 21 detects the relative positions of the diffraction grating 41 and the diffraction grating 42 which will be described later. Further, the control unit 12 or the like also functions as a measurement unit configured to measure the diffracted light from the mold mark 10 and the wafer mark 11 and measure the relative position of the wafer 8 . The illumination optical system 22 guides the light from the alignment light source 23 onto the same optical axis as the detection optical system 21 using a prism 24 or the like, and illuminates the mold mark 10 and the wafer mark 11 . It constitutes a part of the constituted lighting unit. For example, a halogen lamp, LED, semiconductor laser (LD), high-pressure mercury lamp, or metal halide lamp configured to irradiate visible or infrared light that does not contain ultraviolet light to cure the resin 9 is the alignment light source 23 . is used as Further, the alignment light source 23 constitutes a part of the lighting unit. The detection optical system 21 and the illumination optical system 22 are configured to share a part of an optical member constituting them, and the prism 24 is disposed on or near the pupil surface of the detection optical system 21 and the illumination optical system 22 . do.

Each of the mold mark 10 and the wafer mark 11 (alignment mark) is constituted by a diffraction grating and has a periodic pattern. The detection optical system 21 captures an image of an interference fringe (moire fringe) generated by interference between the rays of the diffracted light from the wafer mark 11 and the mold mark 10 illuminated by the illumination optical system 22 to the imaging device ( 25) image above. As the image pickup device 25, CCD, CMOS, or the like is used. Since the diffracted light of the mold mark 10 and the wafer mark 11 is used to generate an interference fringe (moire fringe), the amount of light of the obtained moire fringe varies depending on the diffraction efficiency of the mold 7 and the wafer 8. . In particular, since the diffraction efficiency periodically changes with respect to a wavelength, a wavelength capable of efficiently detecting a moire fringe and a wavelength at which it is difficult to detect a moire fringe appear. Light of a wavelength for which it is difficult to detect a moire fringe may become noise.

On the bonding surface of the prism 24, a reflective film 24a for reflecting the light of the peripheral portion of the pupil surface of the illumination optical system is constituted. The reflective film 24a also functions as an aperture stop that defines the size of the pupil of the detection optical system 21 (or detection NA:NAo). Here, the prism 24 may be a half prism having a semi-transparent film on the bonding surface, a plate-shaped optical element having a reflective film on the outer surface, etc., not limited to the prism. Also, a configuration in which the peripheral portion of the prism 24 of FIG. 3 functions as a transmitting portion, and the central portion functions as a reflecting portion, and the positions of the alignment light source 23 and the image pickup device 25 are exchanged can be applied.

In addition, the position at which the prism 24 according to the first embodiment is disposed does not necessarily need to be on or near the pupil surfaces of the detection optical system 21 and the illumination optical system 22 . 4 is a diagram showing an example of the configuration of such a measurement optical system 3 . In the configuration of the measurement optical system 3 in this case, as shown in FIG. 4 , the detection optical system 21 and the illumination optical system 22 have aperture stop 26 and 27 on the pupil surface, respectively. In addition, as the prism 24, a half prism or the like having a semi-transparent film on its bonding surface is used.

1 is a diagram showing details of an alignment light source 23 according to the first embodiment. FIG. 1A is a schematic diagram showing an example of the alignment light source 23 , and FIG. 1B is a schematic diagram showing an example of the fiber end face 32 . In the alignment light source 23 in Fig. 1A, for example, a semiconductor laser is configured as light sources 30a to 30g which are a plurality of light emitting elements. In addition, the light sources 30a to 30g are not limited to semiconductor lasers. An LED may be used, or a lamp such as a halogen lamp, a metal halide lamp, a high pressure mercury lamp, or a sodium lamp may be used. Also, each of the light sources may be present in a mixed manner. The wavelength of the semiconductor laser configured as the light sources 30a to 30g may be different in all the light sources 30a to 30g, or some may have the same wavelength in order to increase the amount of light of a specific wavelength. However, the light sources 30a to 30g have at least two or more wavelengths. For example, the light source has a first wavelength and a second wavelength that is a different wavelength band from the first wavelength. Also, although seven light sources 30a to 30g are illustrated in FIG. 1A , the number is not limited to seven. At least one light source may be provided, and any number of light sources may be applied. It is desirable to have many alternatives to the choice of wavelength in order to improve the measurement precision. In the case of a semiconductor laser, the semiconductor laser of each wavelength can be individually turned on and off and the amount of light can be adjusted.

The branched fiber (7-to-1 branched fiber) 31 is composed of a plurality of fiber wires in which one end is separated and the other end is bundled. One end is connected to the plurality of light sources 30a to 30g and the other end is connected to the optical rod 33 . In Embodiment 1, the branch fiber is composed of 7 fiber wires. However, the number is not limited thereto, and any number may be applied. In the fiber end face 32 in FIG. 1B, the shaded portion represents the core of the fiber. For example, when the core diameter of each fiber is ?0.4mm, the diameter of the bundled fiber end face 32 may be approximately ?1.3mm including the clad part. Seven fiber wires are tied so that the area does not expand. However, they may be arranged in a straight line, may be arranged concentrically, or may be arranged to form different shapes. Light rays from the light sources 30a to 30g are each guided into seven fibers and integrated into the fiber end face 32 . Since the semiconductor lasers of the respective wavelengths are synthesized using the branched fibers 31, there are advantages in that the degree of freedom in arrangement of the semiconductor lasers is increased and adjustment at the time of replacement is easy.

The alignment light source 23 includes a lamp light source having a wavelength distribution of a wide wavelength band (broad wavelength band) in order to generate illumination light including a plurality of wavelengths. Further, the alignment light source 23 may include a long-wavelength cut filter for blocking the long-wavelength side of the light generated by the lamp light source and a short-wavelength cut filter for blocking the short-wavelength side of the light generated by the lamp light source. The long wavelength cut filter and the short wavelength cut filter may be wavelength cut filters in which a transmission band continuously changes according to an irradiation position. By using such a wavelength cut filter, a specific wavelength can be transmitted.

The optical rod 33 is an example of an optical integrator that bundles the light emitted from the light sources 30a to 30g and makes the light incident on the optical rod 33 . That is, the optical rod 33 may equalize the spatial light intensity distribution of the light emitted from the optical rod 33 . Also, other optical integrators capable of uniform distribution may be used. For example, a microlens array may be employed.

The synthesis method is not limited to synthesis using fibers (special fibers). For example, a method of synthesizing light rays having different wavelengths using a dichroic mirror may be applied, or synthesizing may be performed using a polarizing beam splitter, a half mirror, or the like. The synthesizing method may be appropriately selected in consideration of the space of the arrangement site, the wavelength of the semiconductor laser, the cost of components, and the like.

Light emitted from the optical rod 33 passes through the ND filter 34, so that the intensity (adjustment of the amount of light) can be adjusted. The ND filter 34 is an element that can adjust the intensity of the light passing therethrough. For example, its transmittance can be adjusted by the type or film thickness of the metal film imparted to the quartz. As for the ND filter 34, for example, in order to adjust the amount of light for the alignment light source 23, a plurality of types of different filters having different transmittances are prepared, and the plurality of filters can be switched according to the required amount of light. can be inserted into the optical path. Alternatively, the ND filter 34 may be a filter whose transmittance continuously changes according to a position through which light transmits. In this case, the transmittance may be adjusted according to the position of the ND filter 34 with respect to the optical path. Further, the light intensity generated by each of the light sources 30a to 30g can be adjusted using, for example, electric current supplied individually to the light sources 30a to 30g. Further, the light intensity generated by each of the light sources 30a to 30g can be adjusted by adjusting the position of the ND filter 34 and can be adjusted by a combination of both.

Light emitted from the ND filter 34 passes through the diffusion plate 35 , and the fiber 36 is irradiated with light. When a semiconductor laser is employed as all or part of the light sources 30a to 30g, the wavelength band of light generated by the semiconductor laser is narrow (several nm). Therefore, noise (speckle noise) may occur in the observed phase due to interference. Here, the observed speckle noise can be reduced by rotating the diffuser plate 35 to temporarily change the state of the waveform. When measurement precision on the order of several nm is required, it is preferable to eliminate speckle noise by configuring the diffuser plate 35 in a rotatable manner. In addition, the method of driving the diffuser plate 35 is not limited to rotation, and the diffuser plate 35 may be shifted or driven in the optical axis direction according to the installation space of the diffuser plate 35 .

5 is a diagram showing an example of the configuration of the alignment light source 23 . The light emitted from the fiber 36 is emitted from the alignment light source 23 as illumination light. In the example of Fig. 1, there is one fiber. However, as shown in Fig. 5, the half mirror 37 can be arranged in the optical path in a dividing (branching) manner, so that the illumination light for the two axes can be obtained by making the light rays incident on the fibers 36a and 36b, respectively. have. In addition, by changing the number of divisions (number of branches), it is possible to obtain illumination light for a plurality of axes without being limited to two axes. The method of splitting the light (the method of splitting the light) is not limited to a half mirror, and by arranging the mirror, a beam of light can be partially split.

FIG. 6 is a diagram showing the relationship between pupil intensity distributions IL1 to IL4 of the illumination optical system 22 of the measurement optical system 3 and the numerical aperture NAO of the detection optical system 21 . In Example 1, the pupil intensity distribution of the illumination optical system 22 includes a first pole IL1, a second pole IL2, a third pole IL3, and a fourth pole IL4. The illumination optical system 22 includes light incident in a direction perpendicular to the direction (first direction) in which the patterns of the mold marks 10 or wafer marks 11 are arranged in the XY plane and light incident parallel to the direction. It is possible to illuminate the mold mark 10 or the wafer mark 11 by this. As described above, by arranging the reflective film 24a functioning as an aperture stop on the pupil surface of the illumination optical system 22 and blocking unnecessary light, a plurality of poles, that is, the first pole IL1 from one alignment light source 23 ) to the fourth pole IL4 may be formed. In this way, when forming a pupil intensity distribution having a plurality of poles, it is not necessary to provide a plurality of light sources. Accordingly, the measurement optical system 3 can be simplified or downsized.

Fig. 7 is a diagram showing an example of an alignment mark for generating a moire fringe. Hereinafter, with reference to FIGS. 7A to 7D , the principle of generating moire by diffracted light from the mold mark 10 and the wafer mark 11 and the relative position of the mold mark 10 and the wafer mark 11 using the moire detection will be described. As shown in FIG. 7 , a diffraction grating (first diffraction grating) 41 provided to the mold 7 as the mold mark 10 and a diffraction grating (second diffraction grating) provided to the wafer 8 as the wafer mark 11 (second diffraction grating) The grating) 42 has a slightly different period of the pattern (lattice) in the measurement direction. When these two diffraction gratings having different grating periods are superimposed, a pattern having a period reflecting the period difference between the diffraction gratings, a so-called moire pattern, is generated by interference between rays of diffracted light from the two diffraction gratings. appear. At this time, the phase of the moire changes according to the relative position of the diffraction grating. Therefore, the relative position between the mold mark 10 and the wafer mark 11, that is, the relative position between the mold 7 and the wafer 8, can be obtained by detecting the moire.

Specifically, when the diffraction grating 41 and the diffraction grating 42 having slightly different periods are superimposed, the rays of the diffracted light from the diffraction gratings 41 and 42 are superimposed. Therefore, as shown in Fig. 7C, moire having a period reflecting the difference between the periods is generated. In the moire, the position of light and dark (phase of the stripes) changes according to the relative positions of the diffraction grating 41 and the diffraction grating 42 . For example, when one of the diffraction gratings 41 and 42 is shifted in the X direction, the moire shown in Fig. 7C changes to the moire shown in Fig. 7D. Moire enlarges the amount of actual position shift between the diffraction grating 41 and the diffraction grating 42 and is generated as a pattern of a large period. Therefore, even when the detection optical system 21 has a low resolution, the relative positions of the diffraction grating 41 and the diffraction grating 42 can be detected with high accuracy.

When the diffraction gratings 41 and 42 are detected in the bright field in order to detect moire, the detection optical system 21 also detects the zero-order light from the diffraction gratings 41 and 42. When the diffraction gratings 41 and 42 are detected in the bright field, the diffraction gratings 41 and 42 are illuminated in the vertical direction and diffracted light diffracted in the vertical direction by the diffraction gratings 41 and 42 is detected. may include The 0th order light can lower the contrast of the moire. Therefore, it is preferable that the metrology optical system 3 has a configuration of a dark field that does not detect zero-order light (that is, the diffraction gratings 41 and 42 are illuminated in an oblique manner). It is a figure which shows an example of the mark for alignment (alignment mark). In Example 1, one of the diffraction gratings 41 and 42 is set as a checkerboard-shaped diffraction grating as shown in Fig. 8A, and the other diffraction grating is Set as a diffraction grating as shown in 8b. The diffraction grating shown in FIG. 8B includes a pattern periodically arranged in a measurement direction (first direction) and a pattern periodically arranged in a direction orthogonal to the measurement direction (second direction).

6, 8A and 8B, light from the first pole IL1 and the second pole IL2 is used to irradiate the diffraction grating (incident on the diffraction grating), and to the checkerboard-shaped diffraction grating. It is diffracted in the Y direction and also diffracted in the X direction. Further, light diffracted in the X direction by the diffraction grating having a slightly different period enters the detection area NAo on the pupil of the detection optical system 21 with relative position information in the X direction and is detected by the image pickup device 25 . . This can be used to determine the relative positions of the two diffraction gratings.

As for the relationship between the pupil intensity distribution shown in Fig. 6 and the diffraction grating shown in Figs. 8A and 8B, the light from the third pole IL3 and the fourth pole IL4 detects the relative positions of the diffraction gratings. not used to However, in the case of detecting the relative positions of the diffraction gratings shown in FIGS. 8C and 8D , the light from the third pole IL3 and the fourth pole IL4 is used for detecting the relative positions of the diffraction gratings, and the first Light from the pole IL1 and the second pole IL2 is not used for detection of the relative position of the diffraction grating. In addition, when the pair of diffraction gratings shown in FIGS. 8A and 8B and the pair of diffraction gratings shown in FIGS. 8C and 8D are disposed within the same field of view of the detection optical system 21 and the relative positions of the two directions are simultaneously detected , the pupil intensity distribution shown in Fig. 6 becomes very effective.

9 is a diagram showing an example of the pattern cross-sectional shape of the mold 7 . Hereinafter, the structure of the cross section of a mark (detection object) is demonstrated as an example with reference to FIGS. 9A and 9B. The mark of the model shown in Fig. 9 has a structure composed of three layers. Since these layers have a step difference, when light is irradiated, light diffraction occurs due to the step difference. For that reason, it can be recognized as a mark. The mark 11a shown in Fig. 9A is a mark in a state having no shape error (manufacturing error). The mark 11b shown in Fig. 9B is a mark in a state having a shape error. A mark 11b shown in Fig. 9B is a mark having an asymmetric shape error (manufacturing error). Since the mark 11a has a symmetrical structure without a step or the like, the diffracted light has an ideal diffraction angle. Therefore, the measured mark position has no error with respect to the actual position. However, since the mark 11b has an asymmetric shape error, the diffraction angle of the diffracted light is deviated from the ideal state by the asymmetric portion. Therefore, the measured mark position has an error (shift amount) with respect to the actual position. If this error is large, there is a possibility that the overlapping precision between the layers (each layer) of the wafer is deteriorated to cause manufacturing defects.

Fig. 10 is a diagram showing an example of a simulation model according to Example 1 and its results. Hereinafter, with reference to FIGS. 10A and 10B, the relationship between the wavelength of the illumination light of a position measurement apparatus and the measurement error of a position measurement apparatus is demonstrated as an example. 10A shows an example of a schematic diagram (cross-sectional shapes of the mold mark 10 and the wafer mark 11) of the simulation model. Also, FIG. 10A shows a state in which imprint processing is being performed for the resin 9 . Fig. 10B shows an example of a simulation result based on the schematic diagram shown in Fig. 10A as an example. Through this simulation, a measured value (wavelength characteristic) representing the relationship between the wavelength of the illumination light of the position measuring device and the measurement error of the position measuring device can be obtained. Here, the measurement error of the position measuring device is a measurement error caused by the wafer mark 11 having an asymmetric shape.

In the simulation shown in FIG. 10, the wafer mark 11 is configured to have a periodic pattern, and the step 52 of the periodic pattern is 200 nm and the pitch 51 is 1,000 nm, whereas the amount of error in the shape ( 53) is 40 nm, and the measured value is calculated. Here, as for the material of each layer, the base material of the wafer 8 is composed of a silicon substrate, the wafer mark 11 is composed of SiO _2. A resin 9 is applied to a portion on the wafer mark 11 , and a mold 7 is placed with the resin 9 interposed therebetween.

In the graph of the simulation result shown in Fig. 10B, the abscissa axis represents the wavelength for illumination performed by the position measuring device, and the ordinate axis represents the amount of measurement error due to the wafer mark 11 having processing asymmetry (measurement). position deception). In addition, in FIG. 10B , measured values, which are results obtained individually through simulation at wavelengths, are shown as approximate curves. Also, as shown in the approximate curve of FIG. 10B , it can be seen that the amount of measurement error varies according to each wavelength, and increases and decreases repeatedly. Therefore, the measurement error can be reduced by selecting a plurality of wavelengths (at least two or more) used for alignment. The simulation model shown in FIGS. 10A and 10B is an example. However, the measurement error of the position measuring device varies depending on the wavelength. That is, when there is an error having an asymmetric shape in the wafer mark 11, the measured value is periodically shifted according to the wavelength. Also, this simulation is done in a dark field. However, even when the simulation is performed in a bright field, the measurement error of the position measuring device varies depending on the wavelength of the illumination light.

11 is a flowchart showing an operation sequence of the imprint apparatus 1 including the position measuring apparatus according to the first embodiment. Hereinafter, with reference to FIG. 11 , a method for determining the optimum wavelength by independently changing the light intensity ratio between at least two or more wavelengths constituting the illumination light from the alignment light source 23 according to the first embodiment will be described. The operation of the imprint apparatus 1 including the position measuring device in Fig. 11 is controlled by the control unit 12 (control unit) executing a computer program.

First, in step S101, a wafer (a substrate used for manufacture of an article) 8 and a mold 7 having a laminate structure used for production are conveyed into the imprint apparatus and then held by a wafer holding unit. Further, in step S101 , the mold 7 for manufacturing the article is conveyed to the mold holding unit 4 by the driving mechanism of the mold holding unit 4 and held by the mold holding unit 4 . . Then, in step S102, the pattern formation area of the wafer 8 and the pattern area of the mold 7 are pre-aligned. For example, the pre-alignment can be done by a position measuring device that measures the relative positions of the mold mark 10 and the wafer mark 11 . Next, in step S103 , the resin 9 (imprint material or resist) is applied (arranged) to the pattern formation region of the wafer 8 by the application unit 6 . Further, either the driving mechanism of the mold holding unit 4 or the wafer stage 5 is driven so that the mold 7 is in contact with the resin 9 on the wafer 8 , and the resin 9 on the wafer 8 is driven. ) in contact with a predetermined pattern of the mold 7 . Thereafter, the wafer 8 is exposed to light (illuminated with light) to cure the resin 9 .

Next, in step S104, in a state in which the predetermined pattern of the mold 7 and the resin 9 are brought into contact with each other, the condition of the optimum wavelength is set. Specifically, the control unit 12 controls the alignment light source 23 and switches the plurality of wavelengths so that the wafer 8 is illuminated with respective rays of illumination light having a plurality of wavelengths. Moreover, the relative positions of the mold mark 10 and the wafer mark 11 are measured using the conditions of a plurality of wavelengths, ie, respective rays of illumination light having wavelengths, and information of the relative positions is acquired. The control unit 12 obtains a measured value based on the information of the relative position obtained by the position measuring device for each of the rays of the illumination light having a plurality of wavelengths. The measured value is data representing the relationship between the wavelength of the illumination light and the measurement error of the position measuring device, and is similar to the measured value illustrated in FIG. 10B as an example.

Next, in step S105, the control unit 12, based on the measured value obtained in step S104, generates an optimum wavelength ( lighting conditions). The optimum wavelength is the intensity of each light beam having a plurality of wavelengths constituting the illumination light generated by the alignment light source 23 . The wafer mark 11 has an asymmetric shape that varies depending on the position due to a manufacturing error. Also, therefore, the measurement error measured by the position measuring device varies depending on the position of the wafer 8 . The optimum wavelength is determined so that this measurement error is reduced (reduced). A method of determining this optimal wavelength will be described below. Next, in step S106, the control unit 12 saves the information of the optimum wavelength of step S105 in a storage unit (not shown). Thereby, when manufacturing an article using the mold 7, it is possible to reduce the measurement error in the position measuring device caused by the wafer mark 11 having an asymmetric shape due to the manufacturing error, and the mold mark 10 ) and the relative position (position information) between the wafer mark 11 can be obtained.

Here, the method of setting the condition of the optimum wavelength in step S104 will be described in detail. In setting conditions, the relative positions of the mold mark 10 and the wafer mark 11 are measured for each of a plurality of wavelengths. First, the case where the alignment light source 23 is a semiconductor laser will be described below as an example. In this case, the wavelength can be switched by turning on and off the current value of the semiconductor laser of a specific wavelength. Since the rise time of the semiconductor laser before stable oscillation is within a few seconds and is high speed, the measurement time delay is small even when the semiconductor laser is repeatedly turned on and off. Therefore, the processing of step S104 can be completed within a short time. In addition, a specific wavelength can be blocked by using a shutter without turning the semiconductor laser on and off.

Next, the case where the alignment light source 23 is a lamp is described below as an example. In this case, a wide wavelength band (optical wavelength band) is applied, and it is advantageous to use a wavelength cut filter. For example, by changing the combination of the short wavelength cut filter and the long wavelength cut filter, a light source having a desired wavelength band can be obtained. In addition, by using the wavelength cut filter in which the transmission band continuously changes according to the incident position of the light, it is possible to finely control the wavelength of the illumination light generated by the alignment light source 23 . In addition, as the alignment light source 23, a lamp light source having a wide wavelength band of 200 nm or more is used, and as the image pickup device 25, a sensor (RGB sensor) capable of detecting color (RGB) is used. In this case, light of each wavelength can be detected at once without switching the wavelength of the alignment light source 23 . Thereby, the time required for measurement can be shortened. In a typical RGB sensor, there are three wavelengths (red, green, and blue) that can be measured. Also, the imaging device 25 may have an RGB color filter.

Next, a case in which a semiconductor laser is used as the alignment light source 23 and an RGB sensor is used as the image pickup device 25 will be described below as an example. In this case, the measurement time can be shortened by detecting the wavelength of the semiconductor laser for each RGB band in a mixed manner. For example, consider a case where the sensitivity of the image pickup device 25 of the RGB sensor is R: 590 to 720 nm, G: 480 to 600 nm, and B: 400 to 540 nm. In this case, the wavelength of the semiconductor laser generates light in a wavelength band within a range of 400 to 480 nm, which can be detected by a B pixel (a pixel having a filter that passes light in the B band). Further, light in a wavelength band within 540 to 590 nm is generated, which can be detected by a G pixel (a pixel having a filter that passes light in the G band). Further, light of a wavelength band within a range of 600 to 720 nm is generated, which can be detected by an R pixel (a pixel having a filter that passes light of the R band). Since one wavelength can be measured from each wavelength, one wavelength can be measured from each band, that is, three wavelengths in total. Thereby, the time required for position measurement for switching the wavelength for setting the condition of the optimum wavelength in step S104 can be shortened.

In addition, as for a method of shortening the time required for determining the optimal wavelength, measurement of all wavelengths of the alignment light source 23 by selecting a part from the wavelengths of the light source configured as the alignment light source 23, acquiring data, and interpolating error can be estimated. As for the method of interpolation, there is a method of fitting based on a result obtained by simulation based on a model having a structure similar to that of the wafer 8 . For example, when a simulation result as in FIG. 10B is obtained, a function approximating FIG. 10B may be calculated, and the coefficients may be fitted using measured data having the function as an initial function of the fitting function. Thereby, a graph having the same tendency as the simulation result can be created. In addition, the measurement error of the wavelength that is not actually measured is calculated from the graph. There is also a method of performing fitting by setting an initial fitting function based on actual measurement results under similar process conditions in the past without using simulation results. Thereby, even when the number of measurement points is reduced, the error with respect to an actual measurement value can be reduced.

In addition, the measured value may be acquired through simulation as shown in FIG. 10 . The simulation may be executed by the control unit 12 or may be executed using a computer connected to the control unit 12 . In the simulation, it should be considered that a discrepancy (variation) occurs between the measured value obtained through the simulation and the measured value when the asymmetric shape of the mark changes. Such a discrepancy may cause an error in the position information detected under the lighting condition determined based on the measurement value. Here, as to a method of reducing the error caused by such a cause, there is a method of performing simulation of an asymmetric shape expected in a laminated structure in a plurality of models. For example, with respect to the plurality of models, there may be models in which the thickness or the amount of inclination of the material to be laminated is different from each other. Then, the sensitivity of the measurement error with respect to the shape error of each wavelength is calculated|required from the simulation result of several models. The measurement results (sensitivity of measurement position deception) of a plurality of models having an asymmetric shape are summed, and a plurality of wavelengths with low sensitivity to shape errors are selected and used as wavelengths for detection. Thereby, in a plurality of models having an asymmetric shape used for calculation, on average, it is possible to perform an alignment favorable to an asymmetric shape error.

10B is an example of the result of simulation under the premise that there is no alignment error (alignment error between the mold 7 and the wafer 8) in the simulation. Therefore, the measurement error = 0 nm represents the alignment error = 0. Also, in actual measurement, a measurement error occurs in the relative position obtained through the pre-alignment in step S102 due to the asymmetric shape of the wafer mark 11 . Accordingly, an alignment error may also exist between the mold 7 and the pattern forming area of the wafer 8 aligned based on the relative positions of the mold mark 10 and the wafer mark 11 detected by the position measuring device. If there is an alignment error, when the relative positions of the mold mark 10 and the wafer mark 11 are detected using a plurality of wavelengths in that state, the result of adding an offset amount corresponding to the alignment error with respect to the graph of FIG. 10B is obtained. can be obtained

Here, in order to obtain the offset amount from the result (measured value) measured by the position measuring device, it is necessary to know the correct relative position (alignment error) between the mold mark 10 and the wafer mark 11 . Therefore, the overlapping state of the wafers 8 after imprinting is inspected using an inspection apparatus (evaluation apparatus) such as an overlap inspection apparatus, and the overlapping error is measured and acquired. The error amount obtained at this time is used as the overlap error of the mold 7 and the wafer 8 in the pre-alignment, and the amount of the measurement error obtained through position measurement at step 104 is evaluated. The control unit 12 acquires the measured value in which the offset amount was corrected based on the measured value obtained by the said method, and the overlap error obtained using the overlap inspection apparatus etc. The corrected measured value may be equivalent to the simulation result shown in FIG. 10B if the simulation result is accurate. That is, the position at which the measurement error of Fig. 10B becomes 0 nm is the result measured by the overlap inspection apparatus.

For example, when the imprint operation in step S103 is performed at a position where the mold 7 and the wafer 8 are shifted by 100 nm, the position of measurement error = 0 nm shown in Fig. 10B is actually a position offset by 100 nm is measured as When the wafer 8 is measured (evaluated) by an overlap inspection apparatus or the like, the overlap error of the wafer 8 is measured to be 100 nm. After this measurement, the measured value is input to a computer in the apparatus as an overlap error at the time of pre-alignment in step S102. Thereafter, it is removed and evaluation is performed.

Next, a method for determining the optimum wavelength based on the evaluation in step S105 will be described below. As a measurement value shown in FIG. 10B as an example, the measurement error varies according to the wavelength of the illumination light. The control unit 12 determines the lighting conditions for the alignment light source 23 of the position measurement device so that a measurement error (error of position information detected by the position measurement device) of the position measurement device is reduced. Specifically, the intensity of each light beam having a plurality of wavelengths constituting the illumination light generated by the alignment light source 23 is determined by adjusting the ratio. Here, the alignment light source 23 is configured such that the intensity of each light beam having a plurality of wavelengths can be adjusted to the light intensity of at least two wavelengths. The alignment light source 23 may be configured such that the intensity can be continuously adjusted for at least one of the light rays having a plurality of wavelengths. As an example, the control unit 12 determines the weight of the light of the light beam having a plurality of wavelengths constituting the illumination light generated by the alignment light source 23 . Here, for example, let n be the number of a plurality of wavelengths of light constituting the illumination light, let each wavelength be λ1, λ2, ...λn, and measurement errors using each wavelength are m1, m2, ... It is called mn, and the coefficients of the weights are called k1, k2, ...kn. The measurement error m after weighting is expressed by the following formula (1).

m=k1×m1+k2×m2+...+kn×mn (1)

In the above formula (1), k1+k2+...+kn=1 holds.

The control unit 12 determines the values of k1 to kn such that m=0 in the above formula (1). By synthesizing the light rays of the wavelengths λ1 to λn at the determined ratios k1 to kn, it is possible to eliminate the measurement error due to the wavelength. When the signs of the measurement errors m1 to mn are all the same, the measurement error m cannot be set to zero by adding them. Therefore, it is necessary to offset and record a specific measurement error and to remove it from the measurement error.

Here, when the control unit 12 has a wavelength λm of zero in any of the measurement errors m1 to mn, by setting 1 to the coefficient km of that wavelength and 0 to the other coefficients, Measurement error due to wavelength can be eliminated. However, as shown in Fig. 10B, since the wavelength at which the measurement error becomes 0 is the wavelength at which the change amount (differential value) increases, it is considered that the sensitivity to the shape error increases. Therefore, when the position measuring device aligns one wafer using the illumination light composed only of light rays of the same wavelength, if there is a variation in the shape error for each pattern formation area of the wafer 8, the measurement for each pattern formation area The error may be large.

Here, the control unit 12 performs weighting by the following method. First, the control unit 12 calculates the sensitivity of the measurement error to the wavelength. Then, the control unit 12 preferentially selects (uses) a wavelength with low sensitivity. Then, the control unit 12 selects at least two types of wavelengths having different signs of the measurement errors m1 to mn. When selecting a wavelength, it is possible to select wavelengths having different signs of measurement errors. At this time, when the amount of light is insufficient, the control unit 12 can increase the amount of light by mixing and using three or more types of wavelengths. In addition, at least one type of wavelength can be selected. In addition, the control unit 12 adjusts and changes the light intensity ratio (light amount ratio) of each wavelength so that the mark can be measured in consideration of the balance with the light amount of other marks measured at the same time.

A method of determining an optimal wavelength based on the evaluation of FIG. 11 , specifically, as an example of each light beam having a plurality of wavelengths constituting the illumination light generated by the alignment light source 23 based on the measurement value shown in FIG. 10B . A method of determining the strength is described below. Here, as an example, it is considered to generate illumination light by synthesizing a wavelength of 600 nm and a wavelength of 680 nm at a light quantity ratio of 1:3 based on the wavelength characteristic shown in Fig. 10B as an example. As an example, FIG. 12 shows the sensor of the position measuring device when the mold mark 10 and the wafer mark 11 are illuminated with illumination light obtained by synthesizing light beams having a wavelength of 600 nm and a wavelength of 680 nm in a light quantity ratio of 1:3. The waveform obtained by At this time, the waveform is a measurement waveform of a moire pattern obtained by slicing an image along a line parallel to the measurement direction. In Fig. 12, the vertical axis indicates the amount of light, and the horizontal axis indicates the position on the sensor. Since the measurement error at the wavelength of 600 nm is 120 nm and the measurement error at the wavelength of 680 nm is -40 nm, when they use illumination light synthesized in a light quantity ratio of 1:3, the sum of the measurement errors is 0 nm (120 nm - 40 nm × 3). Therefore, it is not necessary to remove the offset. Further, since the wavelengths of 600 nm and 680 nm are insensitive to the amount of change in the step 52, measurement with high precision (the mold mark 10 and the wafer mark 11) regardless of the change in the amount of change in the step 52. relative position).

A method of performing the synthesis in consideration of the intensity ratio of the wavelengths of the plurality of light rays will be described below as an example in which the alignment light source 23 is a plurality of semiconductor lasers. In this case, the method of adjusting the output light amount by adjusting the driving current value of each semiconductor laser can be applied. Further, as for the other method, a method of disposing the ND filter 34 and changing the transmittance of the ND filter 34 before synthesizing the optical path of the semiconductor laser can be applied.

A case in which the alignment light source 23 is a light source such as a halogen lamp having a wide wavelength band will be described below as an example. In this case, it is difficult to take out only a wavelength of 600 nm and a wavelength of 680 nm from one light source and guide the light. Here, FIG. 13 shows an example of the configuration of the light source 30 constituting the alignment light source 23 . In the example of FIG. 13A , the alignment light source 23 includes lamps 60a and 60b (halogen lamps, etc.), and the short wavelength cut filters 61a and 61b are the long wavelength cut filters 62a and 62b and the half mirror 63. includes Light rays from lamps 60a and 60b each pass through wavelength cut filters having different cutoff wavelengths and are combined by half mirror 63 . Thereby, for example, when light having a wavelength of 600 nm is extracted from the light generated by the lamp 60a and light having a wavelength of 680 nm is extracted from the light generated by the lamp 60b, the Light rays can be synthesized to create illumination light.

The amount of light at each wavelength can be adjusted by adjusting the voltage applied to the lamps 60a and 60b or by performing a method of arranging the ND filter 34 in the optical path before synthesis is performed. When half mirror 63 is used, half of the light from lamps 60a and 60b is transmitted and the other half is reflected. Therefore, it is efficient to supply as many as two scopes of light. In the example of Fig. 13B, the light from the lamp 60 (halogen lamp, etc.) is separated by the half mirror 63a, and wavelength cut filters 61a, 62a, 61b, and 62b are arranged in respective optical paths. The ND filter 34 may be used to adjust the amount of light of each wavelength. The light beams dimmed by the ND filter 34 may be synthesized by the half mirror 63b to generate illumination light.

In addition, it is possible to reduce the overall measurement error by selecting the wavelength range so that it is a wide wavelength band and the measurement error is 0 nm when integrated. For example, in FIG. 10B , integration of a measurement error within a wavelength range of 610 nm to 700 nm results in 0 nm. Therefore, when the illumination light of such a range is used, the offset of a measurement error does not generate|occur|produce. In addition, since not only the wavelength of 640 nm, which is sensitive to the measurement error with respect to the amount of change in the step 52 but also the insensitive wavelength, are used, the sensitivity of the measurement error to the amount of change in the step 52 is reduced.

Hereinafter, an example in which a light source having a wide wavelength band is used as the alignment light source 23 and a color (RGB) sensor is used as the image pickup device 25 will be described. In this case, the positions of the moire fringes detected in the R, G, and B wavelength bands may be detected. Therefore, even without changing the wavelength band of the alignment light source 23, by weighting and averaging the moire fringe positions detected in the R, G, and B wavelength bands on the sensor side, the measurement error due to the wavelength is eliminated. can be done Specifically, the halogen lamp outputs light within a wavelength range of 400 to 720 nm to the alignment light source 23 . A case will be described in which the sensitivity of the color (RGB) sensor of the image pickup device 25 is set to R: 590 to 720 nm, G: 480 to 600 nm, and B: 400 to 540 nm. Since the halogen lamp has a wide wavelength band, waveforms are detected in the bands of R, G, and B of the color (RGB) sensor of the image pickup device 25 . Since the wavelengths are different from each other, the detected positions are different from each other. In addition, measurement errors can be reduced by weighting and averaging the wavelength bands of R, G, and B so as to be similar to the measurement results in an overlap inspection apparatus or the like. Also, this method uses the results of simulations. However, the measurement error can be reliably reduced by using the measured values at each wavelength.

Here, an example in which laser light sources (660 nm, 730 nm, and 760 nm) of three types (three regions) are used as the light source (light emitting element) 30 in the alignment light source 23 will be described below. As for the kind of wavelength of the laser, if the wavelength is dense, measurement can be performed with high precision, but the number of lasers can be determined in consideration of the space and cost for the alignment light source 23 . Also, regarding the type of laser, since there is a difference between the emission intensity due to the wavelength, a laser that meets the required amount of light can be selected. For example, when an output of 1W or more is required, the type of wavelength is limited. For the three wavelengths (660 nm, 730 nm, and 760 nm), a laser having a small size and high power is selected.

Compared with a method of cutting out wavelengths of three regions from a light source having a wide wavelength band, such as a halogen lamp, the use of a laser light source can increase light output. For example, in the case of a halogen lamp or a metal halide lamp, since the light spreads out from the light emitting point, it is difficult to efficiently condense the light to a small area of ?3 mm or less. In order to align the semiconductor wafer, it is necessary to irradiate a region of ?3 mm or less on the wafer 8 with light to observe the mark position. Therefore, the light that cannot be focused on the mark area on the wafer 8 cannot be used for measurement and becomes unnecessary light. That is, when a halogen lamp or the like is used to condense light in a very small area, the efficiency deteriorates. When a laser light source is applied, an area of Φ1 mm or less can be irradiated with high luminance. When the wafer mark 11 on the wafer 8 can be irradiated with high luminance, the reflected light from the wafer mark 11 is detected even when the material laminated on the wafer 8 in the process is a material that is difficult to transmit light. can do.

In addition, when the wavelength of one type of alignment light source 23 is applied, the wavelength becomes a wavelength that is difficult to transmit depending on the material laminated on the wafer 8 or its thickness, and the wafer mark 11 on the wafer 8 is may become undetectable. Therefore, a laser having three types of wavelengths is applied. By synthesizing and using the light beams from each laser, the possibility that the wafer mark 11 cannot be detected due to the stacked structure of the wafer 8 can be reduced.

A method of selecting an optimal wavelength when using three types of wavelengths as described above will be described below based on a flowchart showing an operation sequence of the imprint apparatus 1 including the position measuring apparatus of FIG. 11 . Here, the measurement value obtained by performing measurement using a position measurement apparatus is compared with the evaluation result obtained using evaluation apparatuses, such as an overlap inspection apparatus, and computes a difference. The calculated difference value becomes a measured value that varies depending on the respective wavelengths as shown in Fig. 10B according to the conditions of the mold mark 10 and the wafer mark 11 to be measured.

As for the sign of the difference value obtained through measurement, when the sign of the measured value of one of the three types of wavelengths is different from the sign of the measured value obtained at the other two kinds of wavelengths, the other wavelength having a different sign and weighting is performed using one of the remaining wavelengths. By weighting, the wavelength can be adapted to the superposition measurement result. For example, if the measured value of the wavelength of 660 nm is negative, the measured value of the wavelength of 730 nm is positive, and the measured value of the wavelength of 760 nm is negative, then the wavelength of 760 nm and the wavelength of 660 nm (or 760 nm The light output of the wavelength of ) is weighted and averaged. Thereby, it is possible to conform to the superposition measurement result (reference) or to generate an illumination light. About two types of wavelengths with the same sign of a measurement value, which wavelength may be selected, and you may select a wavelength from the intensity ratio with the different mark observed by the same visual field.

Regarding the three kinds of wavelengths, when the sign of the measured value is the same for all three kinds of wavelengths, it is necessary to remove the offset no matter which wavelength is selected. In this case, it is possible to determine the weighting ratio from the light quantity of the marks for rough alignment observed at the same time.

Here, the mark observed from the same visual field is demonstrated in detail. 14 is a diagram schematically showing alignment marks measured when the mold 7 and the wafer 8 are superposed. A range 73 of the outer frame indicates a range that can be observed by the position measuring device at a time. Based on the geometric center position of the mold mark 71a-1 and the wafer mark 72a-1 as a reference, the relative positional deviation of the mold 7 and the wafer 8 from the measurement result of the mark position of the measurement optical system 3 ( D1) can be obtained. The size of the mold mark 71a-1 and the wafer mark 72a-1 can be reduced. Thereby, it is possible to perform coarse alignment using a mark having a small dedicated area. At this time, an intensity ratio occurs in the mark detected by the difference between the reflectances of the mold mark 71a-1 and the wafer mark 72a-1. When the intensity ratio is large, the irradiation is performed with illumination of a luminance capable of detecting a mark with a weak intensity. Therefore, a mark with strong intensity is saturated, and a measurement error occurs. Therefore, it is necessary to suppress the intensity ratio of the mark.

Next, a moire pattern formed by superposing the mold mark 71a-2 and the wafer mark 72a-2 will be described below. The mold mark 71a-2 and the wafer mark 72a-2 are configured to have the periodic pattern shown in Fig. 8C or Fig. 8D. Since the period in the measurement direction is slightly different, when they overlap, a moire fringe is formed in the Y direction. Further, due to the difference between the periods of the mold mark 10 and the wafer mark 11, the shift direction of the moire fringe changes when the relative position changes.

For example, when the period of the mold mark 71a-2 is slightly larger than the period of the wafer mark, when the wafer 8 is relatively shifted in the +Y direction, the moire fringe is also shifted in the +Y direction. On the other hand, when the period of the mold mark 10 is slightly smaller than the period of the wafer mark 11, when the wafer 8 is relatively shifted in the +Y direction, the moire fringe is shifted in the -Y direction. Here, second-level moire fringes 71a-2' and 72a-2' are formed on the mold mark 71a-2' and the wafer mark 72a-2', and the mold mark 71a-2 and the wafer mark The period of the measurement direction of (72a-2) is switched therebetween. Therefore, when the relative position is changed, the position of the measured second-stage moire pattern changes in the opposite direction. Further, the relative positional shift between the mold 7 and the wafer 8 is obtained from the relative positional shift D2. At this time, even when the periodic marks on the mold side that generate the moire signal and the substrate side are shifted by one period, the shift by one period cannot be detected due to the principle of detecting the moire signal. Therefore, it is confirmed that there is no relative positional shift for one cycle between the mold 7 and the wafer 8 using the mold mark 71a-1 and the wafer mark 72a-1 with low measurement accuracy.

The mold mark 71a-1 and the wafer mark 72a-1 generate a moire signal when the periodic marks on the mold 7 side and the wafer 8 side have a pitch that does not cause a position error by one period. It can be a sign that says

Since the mold mark 71a-1 and the wafer mark 72a-1 are made of different materials, the amount of light detected may differ depending on the wavelength. Therefore, the intensity ratio of the mark can be changed by changing the light output of the three types of lasers. Therefore, weighting is applied to the outputs of the three types of lasers so that the intensity ratio of the moire fringes of the mold mark 71a-1 and the wafer mark 72a-1 is within a range that can be measured. Thereby, the moire pattern of the mold mark 71a-1 and the wafer mark 72a-1 can be measured with one measurement.

In addition, the measured value at wavelengths other than the three types of wavelengths actually measured can be inferred by giving the error which has an asymmetric shape assumed to the structure of the mold 7 and the wafer 8, and performing a simulation. Accordingly, it is possible to calculate the sensitivity of each wavelength with respect to the amount of asymmetry error of the mark shape, and it is possible to preferentially use the wavelength whose sensitivity is lowered.

Hereinafter, with reference to FIGS. 15-17, the distribution of the measurement error in the surface of the wafer 8 is demonstrated. 15 to 17 are diagrams showing examples of measurement errors occurring in the surface of the wafer 8 . 15A is a schematic diagram of differences between patterns of marks at wafer locations. In Fig. 15A, the mold mark 10 and the wafer mark 11 are respectively shown at the left edge, the central part, and the right edge of the wafer 8. As shown in FIG. When a solvent such as resin 9 (imprint material or resist) is applied to a part on the wafer 8 through spin coating, film thickness non-uniformity occurs in a manner symmetrical with respect to the center of the wafer 8 . Therefore, depending on the conditions for forming the wafer pattern, an asymmetric processing error gradually occurs in such a way that it becomes symmetric with respect to the center, and thus a large asymmetry occurs at the wafer edge. In this case, the shape of the asymmetric processing error of the pattern in the radial direction of the wafer (radial direction of the wafer 8) becomes large. Therefore, when measurement is performed at a single wavelength, the measurement error increases in the radial direction of the wafer. For example, the radial direction of the wafer is a direction from the center of gravity of the wafer 8 or a direction such as a radial direction of the wafer 8 . 15B shows an example thereof. The abscissa axis represents the radial position of the wafer, and the ordinate axis represents the measurement error with respect to the true position. For example, the true position is measured and obtained using a measuring device or the like outside the reference device. Each of the lines of condition 81 and condition 82 in FIG. 15B represents a measurement error at a different wavelength. In addition, the lines of condition 81 and condition 82 shown in Fig. 15B are examples, and the measurement error is expressed by a linear function in the radial direction of the wafer. However, higher order components may be included.

When measurement is performed at a single wavelength, a measurement error occurs in the radial direction of the wafer. However, measurement errors in the radial direction of the wafer can be reduced or eliminated by measuring a plurality of wavelengths in a mixed manner. For example, taking the case of FIG. 15B as an example, if synthesis is performed at a ratio of 2 (intensity ratio of wavelength in condition 81) to 1 (intensity ratio of wavelength in condition 82), in the radial direction of the wafer The measurement error becomes zero. Therefore, measurement (relative position between the mold mark 10 and the wafer mark 11) can be performed with high precision regardless of a predetermined position in the radial direction of the wafer.

Here, the measurement results obtained in the conditions 81 and 82 are given a predetermined ratio, that is, the ratio of the measurement result of the condition 81 to the measurement result of the condition 82 = 2 (condition (81)): 1 (condition ( 82)), weighted and averaged to get the true position. At this time, if the measurement error is calculated by weighting and averaging the measurement results at a single wavelength individually obtained, it is necessary to measure as many as the number of wavelengths used for one mark, and thus the measurement time becomes long. . Here, by synthesizing the light intensities of wavelengths at an optimal intensity ratio to generate illumination light and irradiating the light to the mark, measurement can be performed at once. Therefore, the measurement time can be shortened. Fig. 15A shows a wafer mark 11 in which the shape of the mark at the center of the wafer 8 is an ideally symmetrical shape. However, in the wafer 8 used as a sample wafer or in normal manufacturing, an asymmetric processing error occurs even at the center of the wafer 8 . 16A shows its schematic diagram. Because there is an asymmetric processing error at the center position of the wafer 8, a measurement error may occur in the measurement result at a single wavelength of the conditions 81 and 82. 16B shows an example of the measured value at this time. In Fig. 16B, an offset occurs at the center of the wafer 8, and the measurement error becomes zero at a position where there is no asymmetric processing error.

As an example, in a model as shown in FIG. 16 , the offset of the measurement error may be reduced or eliminated by synthesizing the wavelengths so that the slope of the measurement error measured at a single wavelength in the radial direction of the wafer is removed. However, the mark structure of the wafer 8 (the structure of the wafer mark 11) used in ordinary manufacturing is not limited to a simple structure in a model as shown in FIG. 17A is a schematic diagram of an example of a model in which an asymmetric processing error occurs at two locations on the wafer 8. As shown in FIG. Fig. 17B shows an example of the measurement result at a single wavelength in the model shown in Fig. 17A. Here, under the conditions 81 and 82 shown in Fig. 17, the ratio of the inclination of the measurement error in the radial direction of the wafer is set to 1:2. In this case, conditions 81 and 82 are weighted in a ratio of 2 (condition (81)): 1 (condition (82)) and averaged so that the inclination of the measurement error in the radial direction of the wafer is eliminated. In this case, an offset occurs in the measurement error. Here, when there are only two conditions such as condition 81 and condition 82, the offset (offset amount) cannot be reduced or eliminated. However, when the wavelength of the condition 83 can be selected, the inclination of the measurement error in the radial direction of the wafer is reduced or eliminated by weighting and averaging the conditions 81, 82, and 83. Not only can it be done, but the offset can also be reduced or eliminated.

Here, when the number of selectable wavelengths is small, it is considered that the inclination of the measurement error in the radial direction of the wafer is reduced or eliminated by the synthesis of the light intensity ratios of the wavelengths, but the offset cannot be reduced or eliminated. Therefore, if the number of selectable wavelengths is large, it is possible to adjust and select the light intensity ratio of the optimum wavelength in relation to the fact that the slope of the measurement error is reduced or eliminated and the offset is reduced or eliminated. In addition, the generated offset is treated as an error that always occurs during measurement of each pattern formation region, and by removing them, measurement (relative position of the mold mark 10 and the wafer mark 11) can be performed with high precision.

Hereinafter, with reference to FIG. 18, in the surface of the wafer 8, the method for acquiring the graph of the measurement error in the radial direction of the wafer at each wavelength is demonstrated. 18 is a diagram showing an example of a layout within the surface of the wafer 8 for measuring measurement errors in the radial direction of the wafer at a plurality of wavelengths. Usually, when exposing the wafer 8, this is performed on a plurality of pattern formation regions in one wafer. Here, in order to obtain a measurement error in the radial direction of the wafer, the measurement error in the radial direction of the wafer can be measured by measuring all the pattern formation regions of one wafer at one predetermined wavelength (specific wavelength). However, in the case of measuring all the pattern formation areas of one wafer at one wavelength using one predetermined wavelength, since the number of wafers 8 is required as much as the wavelength to be measured and the number of wafers 8 to be measured, the shape etc. A difference arises. As a result, errors occur in each wafer. Here, in the case of using the method shown in Fig. 18, the control unit 12 performs mark position measurement (relative positions of the mold mark 10 and the wafer mark 11), thereby causing an error due to the number of wafers 8. and measurement errors of the wafer 8 can be reduced. In FIG. 18 , the horizontal axis indicates the position in the X direction, and the vertical axis indicates the position in the Y direction. The circle W represents the edge of the wafer 8 . A section having a substantially rectangular shape within the wafer surface represents each pattern forming region. Each of the condition 81, 82, and 83 is a condition having a different wavelength, and the difference between the wavelengths is indicated by displaying the pattern formation area measured at each wavelength with different hatching processing. 18 , three wavelengths are shown as an example. That is, by changing the wavelength used for the relative position of the mold mark 10 and the wafer mark 11 according to the pattern formation area, the measurement error with respect to the position in the radial direction of the wafer at a plurality of wavelengths with one wafer is efficiently reduced. can be obtained with

Here, a reference evaluation value is obtained by measuring a measurement error after the wafer 8 is exposed using an external measuring device or the like, and the control unit 12 calculates a measurement error and a difference from the evaluation value. By calculating the difference, each measurement error at a plurality of wavelengths can be obtained. In FIG. 18, the pattern formation region of the wafer 8 is irradiated with three wavelengths having a constant periodicity. However, the wavelength is not limited to three wavelengths and more wavelengths may be used. It is preferable that it is at least two wavelengths or more.

19 shows an example of a case where a predetermined wavelength is concentrated and distributed in the vicinity of the center of the wafer 8 to perform mark position measurement. As shown in Fig. 19, when the mark position measurement is performed by concentrating and distributing one predetermined wavelength only in the vicinity of the center in the surface of the wafer 8, at that wavelength, the distribution of the measured values in the radial direction of the wafer is in the vicinity of the center. There are problems that can only be detected in In FIG. 19 , since the condition 82 is concentrated near the center of the wafer 8, at the wavelength of the condition 82, a measurement error can be obtained only in a small range in the radial direction of the wafer. Therefore, if the measurement error with respect to the true position increases, a large error occurs in the optimal intensity ratio when the light intensity ratio of the wavelength is weighted and averaged. Therefore, in order to reduce the measurement error, as shown in Fig. 18, the control unit 12 preferably acquires the measurement error in the radial direction of the wafer at a plurality of wavelengths.

Further, it is more preferable to randomly distribute a plurality of wavelengths of at least two or more types in each pattern formation region on the entire surface of the wafer 8, and to measure the position of the mark to obtain a measurement error. Specifically, a measurement error is acquired by randomly selecting one type from at least two or more wavelengths for each pattern formation area and performing mark position measurement on a predetermined pattern formation area. In addition, it is possible to obtain a measurement error by periodically dispersing at least two or more wavelengths for each pattern formation region (for each shot region) to perform mark position measurement. Further, it is possible to obtain a measurement error by measuring the mark position by changing the composition of the light rays (illumination light) of the wavelength to be irradiated for each predetermined pattern formation region in the vertical direction or the horizontal direction of the wafer 8 . Here, the predetermined pattern forming area may be a preset pattern forming area or a pattern forming area arbitrarily selected by a user.

20 shows an example of a case where mark position measurement is performed by distributing a predetermined wavelength in a predetermined direction on the horizontal plane of the wafer 8 . The direction in the horizontal plane represents the X direction (horizontal direction) or Y direction (vertical direction) on the surface of the wafer 8 . As shown in FIG. 20 , in addition to the radial direction of the wafer, a specific wavelength distribution may exist in the X direction or the Y direction of the wafer 8 according to the structure of the cross section of the wafer 8 . In Fig. 20, condition 81 has a distribution in the Y direction, and condition 83 has a distribution in the X direction. Here, for example, when the shape pattern of the wafer mark 11 is formed with an asymmetric processing error in the X direction, the inclination of the measurement error occurs in the X direction of the wafer 8 when measurement is performed at a single wavelength. Therefore, it is necessary to determine the light intensity ratio of the wavelengths so that the measurement error in the X direction of the wafer 8 is reduced or eliminated. The Y direction is the same as the X direction. Therefore, for example, even when the measurement error of each wavelength is acquired in the case of the distribution shown in FIG. 20, the measurement error increases. In the condition (81), it is difficult to accurately obtain the distribution of the measurement error in the X direction, and in the condition (83), it is difficult to accurately obtain the distribution of the measurement error in the Y direction. Therefore, the measurement error increases even when weighting and averaging are performed. Therefore, it is necessary to reduce the measurement error by measuring the mark position using the method described above.

In the above example, the measurement error in the radial direction of the wafer is acquired from data of one wafer. However, similar measurements can be made for a plurality of wafers, and the average value can be used. By using the value calculated|required from the average value, the precision of acquisition of the distribution of a measurement error can be improved.

Hereinafter, with reference to FIG. 21, the method of acquiring the measurement error of the radial direction of a wafer is demonstrated. 21 is an example of a sequence diagram for measuring a measurement error within the surface of the wafer 8 at a plurality of wavelengths. Fig. 21A shows the details of the processing in step S104 in Fig. 11, and Fig. 21B shows the details of the processing in step S105 in Fig. 11 . The operation of the imprint apparatus 1 including the position measuring device of FIG. 21 is controlled by the control unit 12 (control unit).

As shown in Fig. 21, measurement errors are respectively acquired at three wavelengths. First, in step S201, a measurement error is performed by measuring the mark position for each pattern formation region in the surface of the wafer 8 using an illumination light of a specific wavelength (wavelength A) by a position measuring device inside the imprint apparatus 1 . to acquire Next, in step S202, a measurement error is performed by measuring the mark position for each pattern formation region in the surface of the wafer 8 using an illumination light of a specific wavelength (wavelength B) by a position measuring device inside the imprint apparatus 1 . to acquire Next, in step S203 , a measurement error is performed by measuring the mark position for each pattern formation region in the surface of the wafer 8 using an illumination light of a specific wavelength (wavelength C) by a position measuring device inside the imprint apparatus 1 . to acquire In FIG. 12, as an example, although mark position measurement was performed with respect to three wavelengths, the number of wavelengths is not limited to this. More than three wavelengths may be used. In this case, similar processing can be performed as many as the number of wavelengths used.

When the processing of steps S201 to S203 is performed, as shown in Fig. 18, it is preferable that a plurality of wavelengths of at least two or more are randomly distributed over the entire surface of the wafer 8 and irradiation is performed. Also, when the wavelength is changed from a specific wavelength to another wavelength, the measurement immediately after the change may become unstable. Therefore, for example, it is preferable to measure the next pattern formation region without changing the wavelength, instead of changing the wavelength for each pattern formation region. Specifically, at a specific wavelength (wavelength A), the measurement error is obtained by measuring the mark position for each pattern formation region measured at the wavelength A in the wafer surface. After the measurement at the wavelength A is completed, at a specific wavelength (wavelength B) other than the wavelength A, the mark position is measured for each pattern formation area where the wavelength B is measured in the area where the wavelength A is not measured to obtain a measurement error . After the measurement at the wavelength B is completed, at a specific wavelength (wavelength C) other than the wavelengths A and B, the mark position measurement is performed on the pattern formation area where the wavelength C is measured in the area where the wavelengths A and B are not measured, Acquire the measurement error.

Next, in step S301, the overlapping state of the wafers 8 is measured (inspected) using an inspection apparatus such as an external overlap inspection apparatus, and the overlapping error is measured and acquired. At this time, the measured value is the true position. Next, in step S302, a measurement error is calculated|required from the difference between the measured value calculated|required by the position measuring apparatus inside the imprint apparatus 1, and the true position calculated|required by the external superimposition inspection apparatus etc. Next, in step S303, the optimum wavelength is calculated based on the slope of the measurement error in the radial direction of the wafer and the amount of offset.

When the optimum wavelength condition is set using two or more wafers 8, the process of FIGS. 21A and 21B is repeated for the number of wafers 8, and the obtained wavelength light intensity ratio is averaged, respectively. The optimum wavelength can be obtained. Here, the light intensity ratio of the optimum wavelength can be obtained from the distribution of the pattern formation area exactly the same as that of the first wafer. In the second wafer, this can be measured as a distribution different from the distribution of the pattern formation area in the first wafer 8 . By averaging the calculation results on the plurality of wafers 8, a more reliable light intensity ratio of wavelengths can be obtained.

Although the example has been described above based on the imprint apparatus 1, this is similarly applied to the semiconductor exposure apparatus. In addition, the detection method using the moire pattern was demonstrated regarding the method of measuring the relative position of the mold 7 and the wafer 8. FIG. However, Example 1 can also be applied to a case where a pattern of a position on a wafer is measured in a bright field and a dark field, or a measurement method for detecting diffracted light from the pattern.

(Embodiment of article manufacturing method)

Next, a method for manufacturing a semiconductor device (semiconductor IC element, liquid crystal display element, MEMS, etc.) using the above-described exposure apparatus (imprint apparatus 1 having a position measuring apparatus) will be described. The semiconductor device has a step of performing position measurement of the wafer 8 and the mold 7 using the above-described exposure apparatus. Further, the semiconductor device has a pattern forming step of forming a pattern on the wafer 8 after performing position measurement. Further, the semiconductor device has at least a processing step of processing the patterned wafer 8 in the pattern forming step and a step of manufacturing an article from the processed substrate. Processing steps include etching, resist stripping, dicing, bonding, packaging, and the like. According to the manufacturing method for manufacturing the present semiconductor device, a semiconductor device of higher quality than that of the prior art can be manufactured.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. The scope of the following claims is to be construed in the broadest possible manner to include all such modifications and equivalent structures and functions.

This application claims the benefit of Japanese Patent Application No. 2020-027874, filed on February 21, 2020, the entirety of which is incorporated herein by reference.

Claims (19)

It is a position measuring device for measuring the position of an object,
an illumination unit configured to illuminate the object with illumination light including light of a first wavelength and light of a second wavelength different from the first wavelength;
a measurement unit configured to measure a position of the object by detecting light from the object illuminated with the illumination light; and
and a control unit configured to adjust a ratio of the light intensity of the first wavelength to the light intensity of the second wavelength such that a measurement error that varies depending on the position of the object in the measurement unit is reduced. According to claim 1,
and the control unit adjusts the adjustment of the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength through weighting. According to claim 1,
and the control unit adjusts the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength so that the measurement error in a direction from the center of gravity of the object is reduced. According to claim 1,
and the control unit adjusts the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength according to a predetermined position of the object so that the measurement error is reduced. According to claim 1,
and the control unit adjusts the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength so that the measurement error in a predetermined direction on the horizontal plane of the object is reduced. According to claim 1,
and the control unit illuminates the object by periodically dispersing the illumination light including the first wavelength and the second wavelength for each pattern formation area of the object. According to claim 1,
and the control unit changes the composition of the illumination light including the first wavelength and the second wavelength illuminated for each predetermined pattern forming area in each of a longitudinal direction or a transverse direction on the object. According to claim 1,
and the control unit randomly selects one from at least two or more wavelengths including the first wavelength and the second wavelength to illuminate the object. According to claim 1,
and the control unit adjusts the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength so that an offset amount of the measurement error is reduced. According to claim 1,
wherein the object includes a predetermined mark. 11. The method of claim 10,
wherein the predetermined mark has a periodic pattern. According to claim 1,
The object includes a substrate,
and the control unit adjusts the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength so that the measurement error in the radial direction of the substrate is reduced. According to claim 1,
The illuminating light is a light in which rays of laser light having different wavelengths are synthesized. According to claim 1,
The illumination light is light transmitted through a filter that transmits a specific wavelength band having a wide wavelength distribution. According to claim 1,
and the control unit adjusts the ratio of the light intensity of the first wavelength to the light intensity of the second wavelength based on a simulation. an overlap inspection device configured to measure a relative position of a substrate and a mold and to illuminate a predetermined mark provided on at least one of the substrate and the mold using an illumination unit of the position measuring device,
an illumination unit configured to illuminate an object with illumination light comprising light of a first wavelength and light of a second wavelength different from the first wavelength;
a measurement unit configured to measure a position of the object by detecting light from the object illuminated with the illumination light; and
and a control unit configured to adjust a ratio of the light intensity of the first wavelength to the light intensity of the second wavelength such that a measurement error that varies depending on the position of the object in the measurement unit is reduced. It is a position measurement method for measuring the position of an object,
an illumination step of illuminating the object with illumination light including light of a first wavelength and light of a second wavelength different from the first wavelength;
a measurement step of measuring a position of the object by detecting light from the object illuminated with the illumination light; and
and an adjustment step of adjusting a ratio of the light intensity of the first wavelength to the light intensity of the second wavelength so that a measurement error that varies depending on the position of the object in the measurement step is reduced. It is an imprint device that transfers a pattern of a mold on a substrate,
an acquisition unit configured to acquire information of a ratio of the light intensity of the first wavelength to the light intensity of the second wavelength, adjusted in advance so that the measurement error in the radial direction of the substrate is reduced;
an illumination unit configured to illuminate an illumination light to a predetermined mark formed on at least one of the substrate and the mold based on the information acquired by the acquisition unit;
a measurement unit configured to measure a relative position of the substrate and the mold by detecting light from the mark illuminated with the illumination light; and
and an alignment unit configured to align the substrate and the mold based on the relative position. A method of manufacturing an article,
A pattern forming step of forming a pattern on a substrate using an imprint apparatus, the imprint apparatus comprising:
an acquisition unit, configured to acquire information of a ratio of the light intensity of the first wavelength to the light intensity of the second wavelength, which has been adjusted in advance so that the measurement error in the radial direction of the substrate is reduced;
an illumination unit configured to illuminate an illumination light on a predetermined mark formed on at least one of the substrate and the mold, based on the information acquired by the acquisition unit;
a measurement unit configured to measure the relative position of the substrate and the mold by detecting light from the mark illuminated with the illumination light; and
a pattern forming step comprising an alignment unit configured to align the substrate and the mold based on the relative position;
a processing step of processing the substrate on which the pattern is formed in the pattern forming step; and
and a manufacturing step of manufacturing an article from the processed substrate.

Images